AI

This is today’s edition of The Download, our weekday newsletter that provides a daily dose of what’s going on in the world of technology. How kids feel about AI, in their own words —Jen Swetzoff and Keeley McNamara, the founding editors of Anyway, an independent print magazine for tweens and teens When we set out to talk to kids about AI, we thought we knew what we’d hear. We expected stories about cheating, inspiring uses, and concerns like deepfakes or job destruction. But what we actually heard from kids aged 10 to 18 had tons of nuance. Many were so deeply against AI or uninterested in making it part of their lives that they didn’t want to talk about it at all. Others worried about cheating, the environmental impact, or AI’s effect on creativity and critical thinking. AI doesn’t yet seem to be something a lot of them are focused on—and they aren’t begging for it, the way they do for iPhones and Snapchat. But what surprised us most was how much young people could teach adults about AI.  Meet the kids navigating an AI-filled future—and hear what they think of it. Scientists just created female clones of male mice Scientists have deliberately turned male mouse embryos into females for the first time. A team based in Japan used a CRISPR-based approach to remove the Y chromosome from male cells and create female clones of male mice.  The feat could change the way scientists think about reproduction, says Takashi Ishiuchi, a reproductive biologist at the University of Yamanashi, who co-led the work. “There’s a fixed concept in our scientific field that we need both females and males for reproduction,” says Ishiuchi. “I think we could change this concept.” He also hopes the technique could help rescue endangered species, particularly in cases where only a few individuals remain. Find out how it works—and what it could mean for the future of reproduction and conservation. —Jessica Hamzelou What’s behind this summer’s heat, and why 2027 could be worse This summer has been a scorcher for much of the Northern Hemisphere. June and July marked the hottest two-month stretch in Europe since record-keeping began, the contiguous US endured its hottest month on record in July, and South Korea saw its highest-ever recorded temperature.   Climate change makes heat waves more likely and more intense. But there’s another factor at play: El Niño, which is already ramping up and is expected to have a bigger effect on global temperatures next year. Here’s what’s behind this summer’s extreme heat—and why 2027 could be even hotter. —Casey Crownhart This story is from The Spark, our weekly climate tech newsletter. Sign up to receive it in your inbox every Wednesday. Inside the “censorship-industrial complex” idea shaping US policy For years, the idea of a “censorship-industrial complex” that suppressed conservative and populist speech spread in right-wing circles online. But now, the theory has made its way into the Trump administration. Over the past nine months, MIT Technology Review investigated its origins and traced its rise.  In a virtual Roundtables session today, senior reporter Eileen Guo and executive editor Amy Nordrum will explore what they discovered, where the theory is going, and what it could mean for the future of democracy and the internet. Register here to join the session at 19:00 GMT / 2:00 pm ET / 11:00 am PT. The must-reads I’ve combed the internet to find you today’s most fun/important/scary/fascinating stories about technology. 1 Trump has authorized businesses to conduct cyberattacks on criminalsVetted firms could use cyber tools against foreign criminal groups. (Reuters $)+ They would operate under the control of the US government. (Bloomberg $)+ Risks include escalation and inadvertent consequences. (Independent) 2 Chinese censorship is leaking into answers from American AIChatGPT, Claude, and Gemini can all echo authoritarian propaganda. (WSJ $)+ Turkey’s new cyber law has stoked fears of a digital police state. (FT $)+ Censorship conspiracy theories have entered the US policy mainstream. (MIT Technology Review)  3 Twitch is mining people’s streams to train Amazon’s AIAll streamers are opted in to train AI with their content by default. (404 Media)+ Users can now opt out, but the announcement has sparked uproar. (BBC)+ AI’s memories are privacy’s next frontier. (MIT Technology Review) 4 Trump has been sued over a service selling faster access to his social postsThe Intercept and the Freedom of the Press Foundation filed the suit. (NPR)+ They call the service “corrupt” and “unconstitutional.” (Ars Technica)+ And say it violates the right of equal access to official information. (Reuters $) 5 Police officers are using Flock cameras to stalk peopleThree Georgia deputies are the latest officers accused of abuse. (NYT $)+ But the surveillance devices still have some supporters. (Atlantic $) 6 AI reporters are now breaking big newsOne just beat human journalists to a story about hacking at OpenAI. (Wired $)+ The milestone has sparked debate about the future of journalism. (Gizmodo) 7 The web’s newest weapon against AI scrapers is a fontShieldFont foils scrapers but keeps pages readable to people. (Ars Technica) 8 Live streaming has turned us all into Perez HiltonIt’s made us both the celebrity and the paparazzi. (New Yorker $) 9 A fossil discovery suggests live birth emerged far earlier than thoughtIt pushes the origin of mammalian live birth back 95 million years. (New Scientist $) 10 Scientific progress may not be slowing after allNew research disputes claims that science is getting less disruptive. (Economist $) Quote of the day “Trump is trying to enrich himself by privatizing government information that he has no right to sell. We won’t let it stand.”  —Ben Muessig, editor-in-chief of The Intercept, explains why his media organization is suing President Trump over his fast-access service to Truth Social posts. One More Thing Puerto Rico’s power struggles Carmen Suárez Vázquez lives just minutes from Puerto Rico’s only coal-fired power plant. Black dust coats her windowpanes and the leaves of the blooming vines around her home. She doesn’t know exactly how the coal pollution got

Quantum computing has alternated between breakthrough darling and overhyped promise in technology circles. Its powerful new capabilities come with a threat to break current cryptography, but for business leaders navigating the noise, the signal should be clear: post-quantum cryptography (PQC) is a manageable evolution, not a crisis. The mathematics behind today’s encrypted digital transactions may yield to quantum computers one day, but the transition to quantum-resistant algorithms is neither sudden nor insurmountable. For executives concerned about disruption, cost, or complexity, a structured and phased approach exists with trusted technology partners like Intel that are already beginning to deliver the infrastructure to make it possible. A natural evolution, not a cliff edge The “quantum threat” narrative often swings between two extremes: imminent catastrophe or distant irrelevance. The reality occupies a more pragmatic middle ground. Quantum computers are highly specialized accelerators that exploit quantum physics to solve specific hard problems. They have the potential to crack modern encryption, but they will not replace classic servers overnight, nor will they instantly break every encryption protocol on the internet. What they will do is gradually shift the security landscape, much as previous cryptographic transitions have done over the past three decades. In late 2024, the Global Risk Institute, a Toronto-based financial services think tank, surveyed 32 quantum computing experts on when a quantum computer could break a 2048-bit RSA key within 24 hours. An average of optimistic and pessimistic estimates from the experts gave it an even 50-50 probability of reaching this code-breaking milestone by 2040. This timeline, uncertain but measurable, creates space for deliberate planning rather than emergency reaction. The near-term focus should be on “harvest now, decrypt later” scenarios, where adversaries collect encrypted data today and then hold it for future decryption later when that capability becomes possible. This is particularly applicable for information requiring confidentiality beyond 10 years. For most enterprises, this can be a manageable risk when addressed through methodical modernization. Government signals as confidence builders The U.S. government has issued new directives for National Security Systems (NSS), which would likely be first on the list for potential quantum attack. Beginning in January 2027, new NSS acquisitions must be capable of supporting Commercial National Security Algorithm Suite 2.0 (CNSA 2.0) requirements for PQC algorithms standardized by the National Institute of Standards and Technology (NIST) and selected by the National Security Agency, the U.S. intelligence agency responsible for signals intelligence and information assurance. Implementation for new systems (with certain exceptions) is then required by 2031, with 100% adoption targeted by 2035. For commercial enterprises, these timelines are not mandates, but could be signposts. They indicate where vendors, standards bodies, and auditors are headed, providing a reference architecture for responsible stewardship. Organizations can borrow this discipline without necessarily copying the exact timelines, using government guidance to calibrate their own risk tolerance and investment cadence. Intel’s role: Infrastructure ready for the transition Intel is at the heart of the AI revolution by delivering quantum-resistant capabilities across our product portfolio. This is not just aspirational roadmap language; it is starting to be shipping technology. For instance, the Intel Xeon 6 Processor already incorporates quantum-safe memory encryption (AES-256) and microcode signing to protect processor integrity. Upcoming platforms will extend post-quantum algorithms to more firmware and software signing, device interconnects, attestations, and secure boot functions, aligning with the most stringent government and industry directives. Post-quantum algorithms carry different key sizes and computational overhead than legacy methods. Intel addresses this through dedicated cryptographic accelerators, optimized libraries, and specialized CPU instructions that reduce latency and preserve service-level agreements. Technologies such as Intel QuickAssist Technology offload cryptographic workloads, enabling enterprises to adopt stronger algorithms without sacrificing performance. PQC is not a processor-alone problem. System builders and application owners must take a comprehensive view spanning solid-state drives, network interface cards, operating systems, hypervisors, applications, and connected services. Intel is delivering its pieces of the stack, while collaborating with ecosystem partners to ensure interoperability and smooth transition paths. A more in-depth discussion of post-quantum algorithms and attacks can be found in my recent blog posted on Intel’s Community forum: “Post-Quantum Crypto: Panic Like It’s 1999?“ A practical roadmap for enterprises The path forward does not require upheaval, just discipline. Organizations can follow a phased approach that mirrors patterns emerging in government and critical infrastructure sectors: Approach PQC as modernization, not mitigation. Frame the transition as an opportunity to strengthen cryptographic foundations, reduce technical debt, and improve system maintainability. Leverage trusted partners. Technology suppliers like Intel are already shipping quantum-resistant capabilities with performance acceleration. Evaluate platform readiness and vendor roadmaps as part of procurement decisions. Start with visibility. Cryptography is embedded throughout modern technology stacks: not just in database encryption settings but in data at rest, data in transit, digital signatures, code signing, device identity, password hashing, and software update mechanisms. Start by mapping where cryptographic assets live, what algorithms protect them, and which data sets have the longest confidentiality requirements. Protect long-lived data first. Not all cryptographic uses age at the same rate. Encryption protecting long-lifespan intellectual property, personal data, or state secrets faces more immediate attention than short-lived session keys or rotating certificates. Focus initial investments on high-value, long-retention data stores and the trust anchors (root certificates, firmware signing keys) that underpin system integrity. Design for evolution and agility. Post-quantum algorithms are not simple drop-in replacements. They carry different key sizes, performance characteristics, and integration requirements that ripple through protocols, APIs, and hardware. Design systems that can transition algorithms without business disruption: testing compatibility, ensuring vendor roadmaps align, and engineering for rotation. The bottom line Quantum computing will reshape cryptography, but despite what occasional click-bait headlines say, it will not upend business overnight. The transition to post-quantum algorithms is a measured, multi-year journey, one that organizations can navigate with confidence by partnering with capable technology providers, prioritizing long-lived data, and designing for agility. Leaders who approach this as an engineering evolution rather than a threat response will not only be ready for whatever timeline quantum delivers; they will emerge

In this article, you will learn the conceptual and practical differences between retrieval and memory in agentic AI systems, and how to combine both effectively….

This is today’s edition of The Download, our weekday newsletter that provides a daily dose of what’s going on in the world of technology. How we picked 35 of the world’s top young scientists and engineers On September 8, MIT Technology Review will reveal its 2026 list of Innovators Under 35, recognizing 35 young people from around the world who are doing groundbreaking scientific work and building clever technical fixes for sticky problems. By finding the top young innovators globally and learning what they’re focused on in their work, we aim to give readers a sense of what advances to expect in the years to come.  As a newsroom, we also use this exercise to help us spot rising talent and get to know some of the best early-career researchers in the fields that we cover. This year, we received 550 nominations. Find out how we whittled them down to 35 of the young innovators shaping the future of technology, and check out last year’s list. —Amy Nordrum How the “censorship-industrial complex” is changing the internet and US policy —Eileen Guo I first heard the term “censorship-industrial complex” on April 15, 2025.  That’s when I got the tip that a small office in the US State Department, which focused on monitoring and countering foreign disinformation from the likes of Russia, Iran, and China, was facing imminent shutdown—the next day.  And the reason? The office was accused of serving as the department’s central hub in the so-called censorship-industrial complex—a sprawling constellation of government agencies, academics, civil society groups, and Big Tech platforms allegedly conspiring to suppress conservative and populist speech online under the guise of combating disinformation.  I broke the story on April 16. But for me, it was just the start of a deep reporting rabbit hole into an idea that had moved from the fringes of the right-wing internet into the Trump administration. For more on what the narrative means for the internet, read my story here. MIT Technology Review Narrated: Montana’s plan to become an experimental medical hub just pushed forward At the end of July, any biotech company in Montana with an experimental drug gained a clear path to selling it to consumers. Companies whose drugs have been through preliminary testing—sometimes in as few as 10 healthy people—can pay $12,500 to apply to a newly established review board. Once approved, they can set their own prices and sell the drugs through experimental treatment clinics, the first of which is likely to open around the end of this year. Montana’s latest right-to-try legislation is unique. While similar laws elsewhere limit access to people with terminal illness, Montana’s system is theoretically open to anyone who gives informed consent and can pay. That includes people desperate for treatments for rare diseases. It also includes those interested in longevity and drugs pitched as preventive therapies. —Jessica Hamzelou This is our latest story to be turned into an MIT Technology Review Narrated podcast, which we publish each week on Spotify and Apple Podcasts. Just navigate to MIT Technology Review Narrated on either platform, and follow us to get all our new content as it’s released. The must-reads I’ve combed the internet to find you today’s most fun/important/scary/fascinating stories about technology. 1 China-linked hackers have hit Taiwan in an “unprecedented” AI attackThey used open-source agents to compromise government websites. (FT $)+ UK military drones were found sending a signal to China. (Cybernews)+ Taiwan’s “silicon shield” could be weakening. (MIT Technology Review) 2 Wall Street firms are paying $100,000 a month to get Trump posts firstTrump Media said more than 10 firms have signed up for the service. (CNN)+ It offers faster access to market-moving posts on Truth Social. (BBC)+ Trump Media also lost $238 million as crypto holdings fell. (CNBC) 3 ICE plans to give officers gloves that can deliver painful electric shocksIt’s set to spend up to $20 million to buy thousands of the devices. (AP News)+ A switch turns them from normal gloves into “electrical mode.” (Guardian) 4 Spotify will label AI artists and stop recommending themThe platform is cracking down on fake performers. (Guardian)+ “AI personas” will appear on artist profiles and track listings. (NYT $) 5 Social media spurred a deadly migrant surge from Morocco to SpainDisinformation encouraged thousands to attempt the crossing. (NYT $) 6 Anthropic’s Claude is adding watermarks to AI text and imagesIt could guarantee votes are counted and kept anonymous. (Axios) 7 Drugs that mimic the brain’s wakefulness signal are taking offOrexin drugs could treat sleep disorders, ADHD and addiction. (Economist $)+ But psychedelics are falling short in clinical trials. (MIT Technology Review) 8 Cargo thieves have turned to violence to steal AI hardwareShipments have disappeared after their escorts were attacked. (Wired $) 9 Scientists may have found the elusive glueball, a particle made of forceA Chinese collider has produced the strongest evidence yet. (Science) 10 A firm selling “100% human-written, never AI” research is entirely AIThe reviewers on the Research Gold site are AI-generated. (404 Media) Quote of the day “I think the fourth wave of slop will be when there’s no longer any meaningful quality hit in slop, when the average piece of slop is better than the best human in that field.”  —Kevin Roose, a technology columnist at The New York Times, tells the Pivot podcast what the next stage of AI slop will look like. One More Thing PATRICK LEGER Are we ready to hand AI agents the keys? We’re starting to give AI agents real autonomy, and we’re not prepared for what could happen next. Any action that can be captured by text is potentially within the purview of AI agents—which is why they can cause so much mischief. “The great paradox of agents is that the very thing that makes them useful—that they’re able to accomplish a range of tasks—involves giving away control,” says Iason Gabriel, a senior staff research scientist at Google DeepMind who focuses on AI ethics. Researchers warn that agents could misinterpret goals, leak sensitive information, fall victim to prompt-injection

Business and technology leaders need no convincing that the time of agentic AI is here. Organizations are rapidly adopting agents, and few executives doubt the technology’s potential to transform work. But many organizations find that realizing the desired return on investment (ROI) from AI hinges on having the right foundation, with inadequate infrastructure and data being major blockers. Agentic AI places considerable new demands on enterprise data systems. The shift from answering questions to taking actions means AI agents need data from across the enterprise, in all its structured and unstructured forms, and with the right business context. To make decisions and act in real time, agents also need frictionless access to the organization’s operational systems—for example, those storing its supply chain, point-of-sale, or human resources data. Legacy data systems, even those updated just a few years ago, struggle to meet these demands. DOWNLOAD THE REPORT As AI agents become embedded more widely in enterprise operations, the need to overcome the restrictions of legacy data systems grows more urgent. If Gartner’s prediction that AI agents will augment or automate 50% of business decisions by 2027 proves correct, organizations must eliminate bottlenecks or risk depriving agents of the data they need to make the right decisions at speed. This report, based on a survey of 300 data and technology executives, explores how legacy systems are limiting the effectiveness of AI agents in many organizations. It finds that a handful of organizations—the data leaders—are having greater success with agentic AI and experiencing fewer data limitations as a result of legacy systems. These leaders offer a guide to creating the right data environment for agents to flourish and trusted systems to scale. Key findings from the report include: Few companies currently provide agentic AI with ample access to enterprise data. Across all the surveyed organizations, AI only has access to an average of 45% of company data. That number falls to 30% or less in organizations categorized as “data laggards”. A select group, however, ensures access to over 70% of their data. These “data leaders” are having greater success with their agents than the rest. Trust in agent decisions is a reflection of data readiness. Today, only around half of surveyed organizations trust that the decisions their AI agents make are accurate and relevant. By contrast, 100% of the data leaders trust their agents’ decisions, a strong indicator that reliable AI requires a reliable data foundation. Data leaders find it easier to achieve agent scale and speed. Two-thirds of data laggards say legacy data systems limit AI agent scaling (66%) and prevent agents from making decisions at speed (68%). Having largely overcome legacy data constraints, the leaders have mostly cleared these roadblocks, with just 8% reporting either constraint. The pressure is on to make data estates agent-ready. Within two years, 100% of respondents plan to be using agentic AI, with 69% expecting to use it widely. Without removing data system constraints, agentic AI will fail to deliver the desired speed and efficiencies it promises. Data access and context are top priorities. The most important initiative to enable scaling among all respondents is improving access to structured and unstructured data for AI agents. Also high on the list is enhancing data and AI governance with business context. Data leaders are also focusing heavily on the automation of data management. Download the full report. This content was produced by Insights, the custom content arm of MIT Technology Review. It was not written by MIT Technology Review’s editorial staff. It was researched, designed, and written by human writers, editors, analysts, and illustrators. This includes the writing of surveys and collection of data for surveys. AI tools that may have been used were limited to secondary production processes that passed thorough human review.